There is currently considerable emphasis on strategies for eradicating HIV from infected individuals. Although much effort is focused on reactivating integrated provirus using small molecules, a second possibility is using broadly neutralizing antibodies (bNAbs) that are highly potent and active against diverse strains and clades of HIV. Recent evidence demonstrating the ability of bNAbs in delaying viral rebound after stopping antiretroviral treatment (ART) indicate that these antibodies may be able to clear or remove cells expressing envelope glycoprotein on their surface, effectively reducing the pool size of persistently-infected cells. We propose to investigate a role for bNAbs in actively eliminating or reducing the size of the persistent viral reservoir, in novel humanized models of ART-suppressed HIV infection that support antibody-effector function. In the first of three specific aims, we will construct `gutted' or helper-dependent adenoviral (HDAd) vectors, encoding the heavy and light chains of three second-generation bNAbs, in view of the downstream intramuscular application in animal models described in Aims 2 and 3. In the second aim, these vectored bNABs will be tested in HIV-infected humanized BLT mice, in which viral loads will be suppressed to undetectable levels by ART, for measurements of the latent viral pool size by genomic DNA qPCR and viral outgrowth assays. In particular, novel humanized mouse models that have active complement and functional natural killer cells will be used to assess the role of bNAb effector function in shrinking the persistently infected reservoir. In the final aim, the vectored bNAbs will be tested for their ability to shrink viral reservoirs in humanized mice derived using peripheral blood leukocytes from ART-suppressed HIV+ patients with undetectable plasma viral loads. On the basis of the investigations to be conducted in the R21 phase of this application, we anticipate establishing a critical role for bNAb-effector activit targeting persistent viral reservoirs. The R33 phase of the application will extend these studies to the testing of tetracycline-regulable bnAbs encoded in HDAds, possible due to large carrying capacity of HdAds (~30 kb). In addition to exploring newer, more potent bNAbs, we will also test the use of the recently described eCD4-Ig, that encodes both soluble CD4-Ig and a CCR5 peptidomimetic and ARM-Hs, which are bivalent small molecules capable of binding the CD4 binding pocket of gp160 and also recruiting anti-dinitrophenol antibodies for effector activity. This approach of using HDAds encoding bNAbs may also be combined with latency reversal agents, in order to clear out infected cells. The results of these investigations should accelerate the use of bNAbs as an adjunct to ART in the treatment of HIV disease, with the potential for functional eradication due to elimination or prevention of reactivation of persistently-infected cells.